A two-tiered mechanism by which Cdc42 controls the localization and activation of an Arp2/3-activating motor complex in yeast.

The establishment of cell polarity in budding yeast involves assembly of actin filaments at specified cortical domains. Elucidation of the underlying mechanism requires an understanding of the machinery that controls actin polymerization and how this machinery is in turn controlled by signaling proteins that respond to polarity cues. We showed previously that the yeast orthologue of the Wiskott-Aldrich Syndrome protein, Bee1/Las17p, and the type I myosins are key regulators of cortical actin polymerization. Here, we demonstrate further that these proteins together with Vrp1p form a multivalent Arp2/3-activating complex. During cell polarization, a bifurcated signaling pathway downstream of the Rho-type GTPase Cdc42p recruits and activates this complex, leading to local assembly of actin filaments. One branch, which requires formin homologues, mediates the recruitment of the Bee1p complex to the cortical site where the activated Cdc42p resides. The other is mediated by the p21-activated kinases, which activate the motor activity of myosin-I through phosphorylation. Together, these findings provide insights into the essential processes leading to polarization of the actin cytoskeleton.

Direct involvement of yeast type I myosins in Cdc42-dependent actin polymerization.

The generation of cortical actin filaments is necessary for processes such as cell motility and cell polarization. Several recent studies have demonstrated that Wiskott-Aldrich syndrome protein (WASP) family proteins and the actin-related protein (Arp) 2/3 complex are key factors in the nucleation of actin filaments in diverse eukaryotic organisms. To identify other factors involved in this process, we have isolated proteins that bind to Bee1p/Las17p, the yeast WASP-like protein, by affinity chromatography and mass spectroscopic analysis. The yeast type I myosins, Myo3p and Myo5p, have both been identified as Bee1p-interacting proteins. Like Bee1p, these myosins are essential for cortical actin assembly as assayed by in vitro reconstitution of actin nucleation sites in permeabilized yeast cells. Analysis using this assay further demonstrated that the motor activity of these myosins is required for the polymerization step, and that actin polymerization depends on phosphorylation of myosin motor domain by p21-activated kinases (PAKs), downstream effectors of the small guanosine triphosphatase, Cdc42p. The type I myosins also interact with the Arp2/3 complex through a sequence at the end of the tail domain homologous to the Arp2/3-activating region of WASP-like proteins. Combined deletions of the Arp2/3-interacting domains of Bee1p and the type I myosins abolish actin nucleation sites at the cortex, suggesting that these proteins function redundantly in the activation of the Arp2/3 complex.

Activation of the yeast Arp2/3 complex by Bee1p, a WASP-family protein.

The Arp2/3 complex is a highly conserved cytoskeletal component that has been implicated in the nucleation of actin filament assembly. Purified Arp2/3 complex has a low intrinsic actin nucleation activity, leading to the hypothesis that an unidentified cellular activator is required for the function of this complex. We showed previously that mutations in the Arp2/3 complex and in Bee1p/Las17p, a member of the Wiskott-Aldrich syndrome protein(WASP) family, lead to a loss of cortical actin structures (patches) in yeast. Bee1p has also been identified as an essential nucleation factor in the reconstitution of actin patches in vitro. Recently, it was reported that WASP-like proteins might interact directly with the Arp2/3 complex through a conserved carboxy-terminal domain. Here, we have shown that Bee1p and the Arp2/3 complex co-immunoprecipitate when expressed at endogenous levels, and that this interaction requires both the Arc15p and Arc19p subunits of the Arp2/3 complex. Furthermore, the carboxy-terminal domain of Bee1p greatly stimulated the nucleation activity of purified Arp2/3 complex in vitro, suggesting a direct role for WASP-family proteins in the activation of the Arp2/3 complex. Interestingly, deletion of the carboxy-terminal domain of Bee1p neither abolished the localization of the Arp2/3 complex, as had been suggested, nor resulted in a severe defect in cortical actin assembly. These results indicate that the function of Bee1p is not mediated entirely through its interaction with the Arp2/3 complex, and that factors redundant with Bee1p might exist to activate the nucleation activity of the Arp2/3 complex.

Identification, chromosomal mapping and tissue-specific expression of hREV3 encoding a putative human DNA polymerase zeta.

The Saccharomyces cerevisiae REV3 gene encodes the catalytic subunit of a non-essential DNA polymerase zeta, which is required for mutagenesis. The rev3 mutants significantly reduce both spontaneous and DNA damage-induced mutation rates. We have identified human cDNA clones from two different libraries whose deduced amino acid sequences bear remarkable homology to the yeast Rev3, and named this gene hREV3. The hREV3 gene was mapped to chromosome 1p32-33 by fluorescence in situ hybridization. The hREV3 encodes an mRNA of >10 kb, and its expression varies in different tissues and appears to be elevated in some but not all of the tumor cell lines we have examined. In light of recent reports of a putative mouse REV3, these results indicate that mammalian cells may also contain a mutagenic pathway which aids in cell survival at the cost of increased mutation.

In vitro reconstitution of cortical actin assembly sites in budding yeast.

We have developed a biochemical approach for identifying the components of cortical actin assembly sites in polarized yeast cells, based on a permeabilized cell assay that we established for actin assembly in vitro. Previous analysis indicated that an activity associated with the cell cortex promotes actin polymerization in the bud. After inactivation by a chemical treatment, this activity can be reconstituted back to the permeabilized cells from a cytoplasmic extract. Fractionation of the extract revealed that the reconstitution depends on two sequentially acting protein factors. Bee1, a cortical actin cytoskeletal protein with sequence homology to Wiskott-Aldrich syndrome protein, is required for the first step of the reconstitution. This finding, together with the severe defects in actin organization associated with the bee1 null mutation, indicates that Bee1 protein plays a direct role in controlling actin polymerization at the cell cortex. The factor that acts in the second step of the reconstitution has been identified by conventional chromatography. It is composed of a novel protein, Pca1. Sequence analysis suggests that Pca1 has the potential to interact with SH3 domain-containing proteins and phospholipids.

(Aryloxy)aryl semicarbazones and related compounds: a novel class of anticonvulsant agents possessing high activity in the maximal electroshock screen.

A number of (aryloxy)aryl semicarbazones and related compounds were synthesized and evaluated for anticonvulsant activities. After intraperitoneal injection to mice, the semicarbazones were examined in the maximal electroshock (MES), subcutaneous pentylenetetrazole (scPTZ), and neurotoxicity (NT) screens. The results indicated that greater protection was obtained in the MES test than the scPTZ screen. Quantitation of approximately one-third of the compounds revealed an average protection index (PI, i.e. TD50/ED50) of approximately 9. After oral administration to rats, a number of compounds displayed significant potencies in the MES screen (ED50 of 1-5 mg/kg) accompanied by very high protection indices. In fact over half the compounds had PI figures of greater than 100, and two were in excess of 300. The compounds were essentially inactive in the scPTZ and NT screens after oral administration to rats. Various compounds displayed greater potencies and PI figures in the mouse intraperitoneal and rat oral screens than three reference clinically used drugs. The data generated supported a binding site hypothesis. Quantitative structure-activity relationships indicated a number of physicochemical parameters which contributed to activity in the MES screen. X-ray crystallography of five compounds suggested the importance of certain interatomic distances and bond angles for activity in the mouse and rat MES screens.